Digital Camera Multi-Exposure Synthesis for Wide Dynamic Range
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital photographing apparatuses struggle to capture images with a wide dynamic range and high grayscale resolution, especially in situations with varying light conditions, leading to narrow dynamic ranges and loss of information in bright and dark areas.
Innovation Solution
A digital photographing apparatus that acquires a reference image at one exposure and additional images at different exposures, using a multi-level threshold map generation unit to classify pixels by brightness, a motion data acquiring unit to identify motion areas, and a weight data unit to synthesize pixels based on motion and exposure data, ensuring accurate alignment and weighting of image data to achieve a wide dynamic range and high grayscale resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image data is acquired at a specific exposure, then the brightness levels and colors of pixels are limited to certain ranges, but the dynamic range of the image becomes narrow and information about subject and grayscale is destroyed
Solution Approach 1:
The patent divides the image acquisition process into multiple exposures (first exposure for reference image, second and third exposures for additional images). This segmentation allows different parts of the image to be captured at different exposure levels, preserving both bright and dark area information that would otherwise be lost in a single exposure.
Solution Approach 2:
The patent changes the exposure parameter by acquiring images at multiple different exposure levels. The reference image is acquired at a first exposure while additional images are acquired at second and third exposures different from the first exposure. This parameter variation enables the system to capture a wider dynamic range and preserve grayscale information across different brightness conditions.
2Loss of information
If multiple images at different exposures are synthesized, then the dynamic range and grayscale resolution are improved, but motion artifacts occur when subjects move between exposures
Solution Approach 1:
The patent performs preliminary classification of pixels into motion and non-motion areas before synthesizing the final image. The motion data acquiring unit classifies pixels of additional images into motion and non-motion areas based on comparison with the reference image. This preliminary classification enables different synthesis strategies to be applied to different pixel types, preventing motion artifacts while preserving dynamic range benefits.
Solution Approach 2:
The patent applies different synthesis weights to different pixel types based on their motion characteristics. Non-motion pixels receive higher weights from additional images to enhance dynamic range, while motion pixels receive lower weights to prevent artifacts. This local differentiation of synthesis quality based on pixel characteristics resolves the contradiction between improving dynamic range and maintaining alignment accuracy.
3Loss of information
If pixels from additional images are weighted highly in synthesis, then the dynamic range is expanded, but motion area pixels introduce artifacts into the final image
Solution Approach 1:
The patent applies different weighting strategies to different pixel types: non-motion pixels are weighted heavily to expand dynamic range, while motion pixels are weighted lightly to minimize artifacts. The first weight data acquiring unit acquires first weight data as a first weight for pixels with first motion data value (non-motion) and a second weight lower than the first weight for pixels with second motion data value (motion). This local quality differentiation resolves the contradiction by allowing dynamic range expansion where safe and artifact prevention where necessary.
Data Source
AI summary
A digital photographing apparatus capable of acquiring data about an image having a wide dynamic range and a high grayscale resolution, a method of controlling the digital photographing apparatus, and a recording medium storing a program to implement the method are provided. An embodiment includes an imaging device that acquires a reference image and additional images at different exposures and a multi-level threshold map generation unit that classifies the pixels of the images into levels according to brightness. The embodiment further includes a motion data acquiring unit that acquires motion data for each pixel based on its respective brightness level and a first weight data acquiring unit that acquires first weight data based on the motion data of each pixel. In addition, the embodiment includes a final image data acquiring unit that synthesizes the pixels of the reference image and the additional images based on first weight data.